Effective cubic anisotropy fields of 100 keV Ne+ implanted layers on (111) garnet films can be directly measured by monitoring the in-plane ac susceptibility or loss while a dc in-plane field is varied along the projection of a cubic hard axis. The effective field is defined by the onset of irreversibility in the in-plane rotation of the magnetization. In the presence of various perpendicular bias fields, the effective field boundary can be plotted for magnetization directions from ∼10° to 90° out-of-plane, with the 90° orientation defining the uniaxial anisotropy field, as reported previously. The effective fields have been measured as a function of temperature for (YSmLuCa)3(FeGe)5O12 films subjected to several implantation dosages and annealing temperatures. The measurement technique is described in an accompanying paper. In a uniform layer, the in-plane effective cubic field should vary as sin ϑM cos2 ϑM, with the magnetization angle given approximately by sin ϑM = Hb/ Hku, where Hb is the applied bias field and Hku is the effective in-plane uniaxial anisotropy field of the layer. The effective-field boundary of a lightly implanted film (8×1013 ions/cm2) behaves generally as expected. The maximum in-plane cubic anisotropy occurs near ϑM = 35°, the direction of a 〈100〉 cubic hard axis and shows relatively little change with annealing, whereas the uniaxial anisotropy field drops rapidly with annealing. The temperature dependence of the uniaxial effective field clearly reflects the variation of the cubic anisotropy. With an implantation dosage of 2×1014/cm2, the cubic hard axes are defined only after annealing. In the as-implanted film, the out-of-plane hard axis directions cannot be located. A heavily implanted film (6×1014 ions/cm2) shows completely anomalous behavior: The cubic hard-axis maximum has totally disappeared, as if the out-of-plane structure has been randomized, although the three-fold symmetry of the in-plane hard directions remains well defined. Annealing has little effect on this behavior, except that after a 600 °C anneal the cubic anisotropy shows a large decrease and the in-plane directions are split off from the original 〈211〉 directions. Similarly, the uniaxial anisotropy gives evidence of a breakup of the layer into regions as described earlier. Long-range diffusion of neon atoms is presumed to be responsible for this behavior. We plan to present a more detailed account of this work at a later date. 1 I. Maartense, C. W. Searle, and H. A. Washburn, J. Appl. Phys. 52, 2361 (1981). 2 I. Maartense, J. Appl. Phys. 53, xxxx (1982).
Read full abstract